The prognosis for State IV children with neuroblastoma is dismal. Aggressive multimodal treatment programs often result in severe malnutrition (PEM) which interferes with therapy and increases morbidity and mortality. The plan of this study is to evaluate the effectiveness of enteral and parenteral nutrition in preventing and/or reversing PEM, maintaining or restoring immune competence, preventing treatment delays, and improving tumor response of 36 children (ages one to six years) with metastatic neuroblastoma receiving chemotherapy (5 drugs: cytoxan, vincristine, DTIC, adriamycin, VM-26) every three weeks. At diagnosis, children who are considered nourished at diagnosis will be randomized to receive either a comprehensive enteral nutrition program or total parenteral nutrition (TPN) for the first 35 days of treatment. Children considered malnourished at diagnosis will be randomized to receive either TPN or partial parenteral nutrition plus enteral nutrition for the first 35 days of the treatment. Thereafter, all children will receive a comprehensive enteral nutrition program. Nutritional status, recall skin test response, treatment delays and tumor response will be evaluated a 0, 6, 12, 18 and 22 weeks. Additional nutritional evaluations (dietary, anthropometric and biochemical) will be determined weekly during the first 5 weeks to identify nutritional indices of clinical and subclinical PEM in children undergoing treatment for metastatic neuroblastoma.